summaryrefslogtreecommitdiff
path: root/locks
Commit message (Expand)AuthorAgeFilesLines
* Thanks to Jeff for pointing out I didn't point to the atomics.wrowe2003-08-081-3/+3
* Revamp apr_thread_mutex to be as thread safe, and occasionallywrowe2003-08-071-23/+57
* Introduce the proc_mutex_no_tryacquire stub, returning APR_ENOTIMPL,wrowe2003-08-061-5/+9
* Invert the order of marking the lock as released. Since we firstwrowe2003-08-061-5/+5
* Clean up a style issue; apr_get_os_error() is much cleaner.wrowe2003-06-181-13/+13
* Win32: Adopt Brian Havard's OS/2 rwlock implementation.stoddard2003-06-181-41/+107
* Error handling fix (unrelated to previous changes): don't calljorton2003-06-081-1/+0
* Code style cleanups: remove unnecessary casts to and from void *;jorton2003-06-081-8/+3
* POSIX says that passing a mutexattr object with default attributes tojorton2003-06-081-18/+1
* Add proc_mutex_lockfile() for retrieving the name of the filetrawick2003-06-075-0/+32
* Don't require the lock file name to be passed intotrawick2003-06-071-0/+3
* When using a temporary file for flock- and fcntl-based mutexes,trawick2003-06-071-7/+4
* Don't segfault trying to close a file in error paths of flocktrawick2003-04-191-3/+6
* We need to call apr_proc_mutex_child_init, whichjim2003-03-301-1/+4
* Bugz 17186jim2003-03-271-0/+1
* Win32 needs to do nothing if the file handle is already open in a childwrowe2003-03-271-18/+51
* Fix two problems in apr_thread_cond_timedwait, the first observed bywrowe2003-02-271-2/+3
* When we generate our own semaphore name (internally) when usingjim2003-02-231-7/+23
* rename apr_arch_fileio.h to apr_arch_file_io.h for consistencythommay2003-01-075-5/+5
* Namespace protection for include/arch/ header filesthommay2003-01-0621-34/+34
* Update copyright notices to 2003.thommay2003-01-0121-21/+21
* A second stab at getting apr_thread_cond_timedwait to work. Someonewrowe2002-12-311-2/+2
* Fix up code in thread_cond. First always escape the case where we outrightwrowe2002-12-291-23/+72
* OS/2: Make apr_proc_mutex_cleanup() public to match recent API change.bjh2002-11-231-3/+3
* Implemention of apr_proc_mutex_cleanup() for NetWarebnicholes2002-11-221-0/+5
* axe the remaining proc_mutex_FOO_destroy() functions, which aretrawick2002-11-211-44/+1
* get apr_proc_mutex_destroy() out from undertrawick2002-11-211-6/+6
* apr_proc_mutex_destroy should be exported.rbb2002-11-211-1/+1
* Register the proc_mutex cleanup with an exported API. Most of thisrbb2002-11-201-23/+18
* Rip out buggy nested-mutex code from apr_proc_mutex. This should fix subtleaaron2002-11-181-58/+3
* Whoa Nelly! Lots of errors in the global mutex failure cases. We havewrowe2002-10-291-17/+32
* Handle leaks right now are a huge problem. Close two of them.wrowe2002-07-311-6/+2
* Collapse this down to the infinately more readable apr_pool_cleanup_run().aaron2002-07-301-6/+1
* When we are destroying a mutex we don't know if it is locked or not,aaron2002-07-301-1/+0
* forewarnedwrowe2002-07-151-1/+5
* Replaced APR_USEC_PER_SEC division with the new apr_time_sec() macrobrianp2002-06-302-3/+5
* Added 2 additional lock functions: apr_proc_mutex_name andjim2002-06-075-5/+67
* Split APR_THREAD_MUTEX_DEFAULT with a new APR_THREAD_MUTEX_UNNESTEDwrowe2002-05-295-13/+85
* Fix a problem with global mutexes on OS/390 when something othertrawick2002-05-091-4/+4
* APR_PROC_MUTEX_IS_GLOBAL is the condition where apr_global_mutex_ttrawick2002-05-081-2/+0
* get global_mutex.c to compile when APR_PROCESS_LOCK_IS_GLOBALtrawick2002-05-061-0/+2
* Add apr_os_get function for global_mutexes.rbb2002-05-021-0/+12
* Optimization: rearranged the mutex lock/unlock code tobrianp2002-04-281-10/+8
* Add .deps to cvsignore since APR may now generate .deps files.jerenkrantz2002-04-223-0/+3
* Allow VPATH builds to properly generate build dependencies. This requiresjerenkrantz2002-04-223-0/+6
* We don't need seperate global_mutex's.dreid2002-04-132-97/+1
* Don't build locks.lo anymore.aaron2002-04-092-2/+2
* ** DEPRECATE old lock API: apr_lock.h **aaron2002-04-099-2753/+2
* Be a bit more clear about the FIXME statement and the exactjim2002-04-052-6/+14
* Support for Posix semaphores for locking has been added. This usesjim2002-04-043-6/+269